For several versions now, I have been getting random crashes that happen intermittently without apparent cause.  Sometimes it lasts a day before crashing, sometimes several.  Ironically, when I used the Fix Common Problems plugin's troubleshooting mode to capture logs where it may happen, it never crashed.  When I turned off Troubleshooting mode, it eventually crashed.  Now that I've updated to 6.7.2, there no longer is a Troubleshooting mode in FCP and it mentions to use the syslog server instead.  I've done that, but the logs show nothing unusual and the system still crashes randomly.  Enclosed are some logs where you can see where it crashed after a day, and another where it went a few days.

 

Any inout appreciated.  Thank you.

syslog-192.168.11.110.log

It's segfaulting. It could be bad RAM or some software incompatiblity. Full diagnostics might reveal more. I'd run a MemTest for 24 hours or so first, then try booting into safe mode.

Hi John.  The segfault does not consistently appear.  Also, if it was bad RAM or software incompatibility, wouldn’t it also crash with the Troubleshooting mode in FCP when it was running?  It never crashed, for whatever reason, as long as that was going so I’m not sure what was different.  Regardless, I’ll try a memtest to narrow things down.

13 minutes ago, arcane said:

Troubleshooting mode in FCP when it was running?

Not necessarily

If you can eliminate the RAM you need to run in safe mode with no plugins, dockers or VMs. Just a vanilla NAS. Once that's stable you can start enabling the extras slowly until you find the incompatible one.

Just an update.  I ran memtest for four days, without any errors.  However, running it in safe mode still does not eliminate the random crashes every few days.  I don't see anything in the log from when it runs normally up until the time it crashes.  I'm at a loss as to what to do now to eliminate the crashes.  Again, not sure why, but when Troubleshooting mode was run in the old FCP, it ran for months without issue.

 

Any suggestions appreciated.  Thank you.

Did you try running without dockers and VMs?

No... just without plugins and I am not using VMs.  So basically just re-enable dockers one at a time?  It's just perplexing why the system was stable until the recent versions, and whatever FCP Troubleshooting did, it restored the stability.  No way to go back to an earlier version since it's deprecated?

Edited November 16, 20196 yr by arcane

My advice is the same as it was nearly a month ago.

 

On 10/21/2019 at 1:46 AM, John_M said:

If you can eliminate the RAM you need to run in safe mode with no plugins, dockers or VMs. Just a vanilla NAS. Once that's stable you can start enabling the extras slowly until you find the incompatible one.

 

You also at some point need to post the diagnostics.  All me know is that ffdetect segfaulted.  We have no clue what is running on the system, what is installed, what your configuration is etc etc etc.

I too have had random reboot problems (see below). I ran a memtest with zero errors, changed the UPS and eventually landed on safe mode. I have been up and running for 10+ days now in safe mode, which is the longest its been up in quite some time. The reason I'm posting in this thread is that I too had the Fix Common Problems plugin installed, and now that I'm thinking back.. I installed it PRETTY close to the first crash of my unraid.. I am going to let mine run in safe mode until after Thanksgiving, if there's no unexpected reboots in that time I will restart it in normal mode and uninstall Fix Common Problems as my first plugin in the process of elimination. Seeing that you're having the same problem and had the same plugin installed around the time it started happening gives me hope.

)

Thanks all for chiming in.  I have the system in safe mode right now with no plugins or Dockers running.  Hopefully it will stay stable for at least a week without crashing.  Enclosed are the diagnostics.

 

ThomasE - I don't know if FCP is causing the crashes.  I think I had FCP installed for a while in previous versions and the system was stable.  It was only when I upgraded unRAID a few versions (don't recall which) ago that the crashes started happening.  This is when I tried to run Troubleshooting mode in FCP to catch the error.  Oddly enough, it never crashed while Troubleshooting mode was on.  But when I rebooted the system to stop Troubleshooting mode, the system again crashed intermittently.  

tower-diagnostics-20191121-0729.zip

Well I started on 6.7.2, maybe FCP was working with an older version. I am still 99% sure it is FCP that's causing my reboots as it was installed right around the time my reboots started happening. I am taking my unraid down this afternoon and booting it into normal mode, then removing FCP. If that works and there are no more reboots, I will post back in here so you can try the same thing and hopefully your reboots will stop as well.